Detailed Process Analysis of Biobased Polybutylene Succinate Microﬁbers Produced by Laboratory-Scale Melt Electrospinning

polymers

Article Detailed Process Analysis of Biobased Polybutylene Succinate Microﬁbers Produced by Laboratory-Scale Melt Electrospinning

Maike-Elisa Ostheller 1,† , Naveen Kumar Balakrishnan 1,† , Robert Groten 2 and Gunnar Seide 1,*

1 Aachen-Maastricht Institute for Biobased Materials (AMIBM), Maastricht University, Brightlands Chemelot Campus, Urmonderbaan 22, 6167 RD Geleen, The Netherlands; [email protected] (M.-E.O.); [email protected] (N.K.B.) 2 Department of Textile and Clothing Technology, Niederrhein University of Applied Sciences, Campus Moenchengladbach, Webschulstrasse 31, 41065 Moenchengladbach, Germany; [email protected] * Correspondence: [email protected] † Both authors contributed equally to this work.

Abstract: Melt electrospinning is widely used to manufacture fibers with diameters in the low micrometer range. Such fibers are suitable for many biomedical applications, including sutures, stents and tissue engineering. We investigated the preparation of polybutylene succinate microfibers using a single-nozzle laboratory-scale device, while varying the electric field strength, process throughput, nozzle-to-collector distance and the temperature of the polymer melt. The formation of a Taylor cone followed by continuous fiber deposition was observed for all process parameters, but whipping behavior was enhanced when the electric field strength was increased from 50 to 60 kV. The narrowest Citation: Ostheller, M.-E.; fibers (30.05 µm) were produced using the following parameters: electric field strength 60 kV, melt Balakrishnan, N.K.; Groten, R.; temperature 235 ◦C, throughput 0.1 mL/min and nozzle-to-collector distance 10 cm. Statistical analy- Seide, G. Detailed Process Analysis of sis confirmed that the electric field strength was the most important parameter controlling the average Biobased Polybutylene Succinate fiber diameter. We therefore report the first production of melt-electrospun polybutylene succinate Microfibers Produced by fibers in the low micrometer range using a laboratory-scale device. This offers an economical and Laboratory-Scale Melt Electrospinning. Polymers 2021, 13, environmentally sustainable alternative to conventional solution electrospinning for the preparation 1024. https://doi.org/10.3390/ of safe fibers in the micrometer range suitable for biomedical applications. polym13071024 Keywords: polybutylene succinate; fiber spinning; nonwoven; environmental sustainability; melt Academic Editor: spinning; fiber production; electrospinning; melt electrospinning; process development; biomedi- Francesco Boschetto cal applications

Received: 8 March 2021 Accepted: 24 March 2021 Published: 26 March 2021 1. Introduction Biocompatible and biodegradable polyesters are increasingly important in biomedical Publisher’s Note: MDPI stays neutral procedures, especially when used as sutures, bone fixation devices, plates, stents and with regard to jurisdictional claims in screws, as well as tissue repair and tissue engineering matrices [1]. Polylactic acid (PLA) published maps and institutional affil- and polybutylene succinate (PBS) are widely available thermoplastic biopolymers that iations. could replace petrochemical polymers in the future [2]. Biobased polyesters for medical applications are typically used in the form of films [3] or scaffolds in tissue engineering [4]. They can be manufactured by salt leaching [5], extrusion [6] or electrospinning [7]. Electrospinning is an efficient method for the manufacture of fibers ranging in diame- Copyright: © 2021 by the authors. ter from a few micrometers to hundreds of nanometers [8–10]. Such fibers are beneficial Licensee MDPI, Basel, Switzerland. because they combine an enormous surface area with high flexibility [11]. They are partic- This article is an open access article ularly useful for tissue engineering because fibers can be spun directly onto the surface distributed under the terms and of another material [12]. Further applications beyond the sphere of biomedicine include conditions of the Creative Commons filtration and separation [13], as well as electronics and energy [14,15]. Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ The global microfibers/nanofibers market was valued at US$477.7 million in 2016 and 4.0/). electrospinning was one of the most widely-used techniques for the production of such

Polymers 2021, 13, 1024. https://doi.org/10.3390/polym13071024 https://www.mdpi.com/journal/polymers Polymers 2021, 13, 1024 2 of 16

fibers [16]. The two major electrospinning methods are solution electrospinning and melt electrospinning, both of which involve the exposure of a liquid to a strong electric field as it flows through a capillary in order to draw out fibers [12]. The large potential difference (tens of kilovolts) applied to the liquid leads to the formation of a jet that undergoes whipping movements, which in turn cause stretching and ultimately the deposition of microscale or nanoscale fibers onto a collector [17]. Solution electrospinning uses polymer substrates dissolved in an organic solvent, which evaporates as the fiber is spun, whereas melt electrospinning uses a high-temperature molten polymer. The lower viscosity and higher electrical conductivity of polymer solutions enables solution electrospinning to produce thinner fibers than melt electrospinning. However, PLA and PBS must be dissolved in toxic solvents such as chloroform and dichloromethane, which can be carried over to the final product. To make the process more environmentally friendly, researchers are currently focusing on the use of more benign solvents such as formic acid and acetic acid [18,19]. Nevertheless, an additional solvent recovery step is required, increasing the overall process costs [12]. For example, the production of 1 kg of PLA fibers, when processed as a 10% solution, requires 10 L of solvent with only 90% recovered in a typical process [20]. In contrast, melt electrospinning does not require solvents, but the melt must be held at high process temperatures in order to facilitate extrusion [21]. Furthermore, the high viscosity and low conductivity of the melt yields fibers with a thicker average diameter than solution electrospinning [11]. Nevertheless, because melt electrospinning has no requirement for solvents, and therefore no need for a solvent recovery step, it is more cost effective and environmentally sustainable than solution electrospinning [22–26]. The wider adoption of melt electrospinning could help to reduce the environmental footprint of current industrial electrospinning processes for medical-grade fibers, but solution electrospinning remains the favored industrial process because it has already been scaled up (and is therefore more economical) and also produces finer fiber products [9]. Solution electrospinning with PBS and PLA has successfully yielded fibers with average diameters in the sub-micrometer range for biomedical applications such as wound dressings. Melt electrospinning has also been carried out with PLA, but the brittleness and low conductivity of the material hinder the production of nanofibers and their applications, and various machine and material modifications have been tested to overcome this [27–37]. Compared to PLA, PBS is more ductile with a lower glass transition temperature (below room temperature). Accordingly, modifications to improve the ductility of this polymer may not be necessary. A porous nonwoven PBS mesh with a low pore size has been produced by solution electrospinning and developed for filtration applications [31]. The evaluation of solution- electrospun microscale and nanoscale PBS fibers for biomedical applications confirmed that their high porosity and hierarchical structure offers sufficient mechanical properties for applications in wound healing and soft tissue engineering [4]. The PBS backbone has more polar functional units than PLA, which means that PBS may be more electrically conductive than PLA in its molten form. But, despite the advantages set out above, melt electrospinning with PBS has, to the best of our knowledge, not been attempted before. Here we report the preparation of the first melt-electrospun PBS microfibers using our single-nozzle laboratory-scale device. We investigated the influence of four different process parameters on fiber diameter: temperature, electric field strength, throughput and nozzle-to-collector distance. We measured the effect of temperature on the viscosity and electrical conductivity of PBS melts. We also examined the thermal properties of the polymer, its susceptibility to degradation during processing and the influence of process parameters on fiber crystallinity. Finally, statistical analysis was used to predict the parameter with the greatest influence on the fiber diameter.

2. Materials and Methods 2.1. Materials Melt electrospinning was carried out using PBS ﬁber-grade resin (FZ78TM) produced by the polymerization of biobased succinic acid and 1,4-butanediol. The manufacturer Polymers 2021, 13, x FOR PEER REVIEW 3 of 18

2. Materials and Methods 2.1. Materials

Polymers 2021, 13, 1024 Melt electrospinning was carried out using PBS fiber-grade resin (FZ78TM) produced3 of 16 by the polymerization of biobased succinic acid and 1,4-butanediol. The manufacturer (MCPP, Düsseldorf, Germany) reported the following specifications: a melt flow rate of 22 g/10 min at 190 °C and using a weight of 2.16 kg, and a crystalline melting temperature of(MCPP, 115 °C. Düsseldorf, The polymer Germany) was vacuum reported dried the at 60 following °C for 12 specifications: h before processing. a melt flow rate of 22 g/10 min at 190 ◦C and using a weight of 2.16 kg, and a crystalline melting temperature ◦ ◦ 2.2.of 115MethodsC. The polymer was vacuum dried at 60 C for 12 h before processing. 2.2.1.2.2. Methods Melt-Electrospinning Equipment 2.2.1.For Melt-Electrospinning the evaluation of general Equipment processability and fiber formation characteristics, we used Forour thelaboratory-scale evaluation of single-fiber general processability melt-electrospinning and fiber device, formation including characteristics, a tempera- we tureused controller, our laboratory-scale high-voltage single-fiber power supply, melt-electrospinning heating elements, device, syringe, including pump a temperatureand collector.controller, The device high-voltage was equipped power with supply, JCS-33A heating temperature elements, process syringe, controllers pump (Shinko and collector. Tech- nos,The deviceOsaka, was Japan) equipped and withPT JCS-33A100 platinum temperature thermocouples process controllers (Omega (Shinko Engineering, Technos, Deckenpfron,Osaka, Japan) Germany) and PT 100 to control platinum the thermocouples melting temperature (Omega (Error! Engineering, Reference Deckenpfron, source not found.Germany)). to control the melting temperature (Figure1).

FigureFigure 1. 1. SchematicSchematic illustration illustration of of the the laboratory-scale laboratory-scale melt-electrospinning melt-electrospinning device. device. In previous studies with PBS, the reported melt-processing temperature was 190 ◦C[38]. OurIn laboratory-scale previous studies melt-electrospinning with PBS, the report deviceed melt-processing does not use an temperature extruder but was uses 190 a glass °C [38].syringe Our instead. laboratory-scale Because we melt-electrospinning are unable to apply device any shear does to not make use the an polymerextruder meltbut uses flow, awe glass started syringe with instead. higher temperaturesBecause we are for theunable trials to to apply compensate. any shear Material to make flowed the polymer from the meltsyringe flow, at 205we ◦startedC but thewith viscosity higher wastemperatur too high.es Onlyfor the temperatures trials to compensate. > 235 ◦C supported Material flowedconstant from and the continuous syringe at fiber 205 formation°C but the withviscosity a Taylor was cone.too high. Therefore, Only temperatures we carried out > 235 the °Ctrials supported with polymer constant melts and atcontinuous 235 and 265 fiber◦C. formation We used with a KNH65 a Taylor high-voltage cone. Therefore, generator we carried(Eltex-Elektrostatik, out the trials Weilwith am polymer Rhein, melts Germany) at 235 with and a 265 range °C. of We 6–60 used kV. a Potential KNH65 differenceshigh-volt- ageof 50 generator and 60 kV (Eltex-Elektrostatik, were applied during Weil the am melt-electrospinning Rhein, Germany) with experiments, a range of with6–60 positivekV. Po- tentialvoltage differences on the collector of 50 and a60 grounded kV were spinneret.applied during The collector the melt-electrospinning was a 6 cm flat aluminum experiments,plate overlaidwith positive with voltage a thin paperboard.on the collecto Differentr and a grounded distances spinneret. between theThe spinneretcollector was and acollector 6 cm flat were aluminum tested, plate such asoverlaid nozzle-to-collector with a thin paperboard. distances of 8Different and 10 cm. distances An 11 Plusbetween spin thepump spinneret (Harvard and Apparatus, collector were Cambridge, tested, such MA, as USA) nozzle-to-collector was used with deliverydistances rates of 8 ofand 0.1 10 or cm.4 mL/min. An 11 Plus A 2-mL spin glasspump syringe (Harvard (Poulten Apparatus, & Graf, Cambridge, Wertheim, MA, Germany) USA) was equipped used with with an additional metal orifice of 0.90 mm served as the spinneret nozzle. The experimental parameters are summarized in Table1.

2.2.2. Polymer Characterization Thermogravimetric analysis (TGA) was carried out using a Q5000 device (TA Instru- ments, Asse, Belgium). We heated 5-mg PBS granules at a rate of 10 ◦C/min under nitrogen ﬂowing at 50 mL/min until the temperature reached 700 ◦C. The temperatures at which 5% and 50% weight loss occurred were determined using TA universal analysis software. Polymers 2021, 13, 1024 4 of 16

Table 1. Experimental parameters used for the melt electrospinning of polybutylene succinate (PBS) ﬁbers.

Temperature (◦C) Electric Field (kV) Nozzle-to-Collector Distance (cm) Throughput (mL/min) 50 8 0.1 50 8 4 50 10 0.1 50 10 4 235 60 8 0.1 60 8 4 60 10 0.1 60 10 4 50 8 0.1 50 8 4 50 10 0.1 50 10 4 265 60 8 0.1 60 8 4 60 10 0.1 60 10 4

The rheological properties of PBS were characterized using a Discovery HR1 hybrid rheometer (TA Instruments) focusing on the frequency-dependent complex viscosity G*. We carried out a frequency sweep from 1 to 628 rad/s. For all experiments, we used a 25-mm plate-to-plate geometry with the distance set to 1000 µm, and the strain amplitude was maintained at 1%. Measurements were taken at temperatures of 145, 175, 205, 235 and 265 ◦C. The complex viscosity of PBS is presented at an angular frequency of 10 rad/s at different temperatures to facilitate comparative analysis. We made the rheological measurements 3 times at each temperature and we have presented the mean values along with the standard deviation for comparison. The electrical resistance of molten PBS was measured at the same temperatures as the rheological properties using a Keithley 617 electrometer (Tektronix, Beaverton, OR, USA). The polymer granules were melted using band heaters, and two electrodes (6 mm apart) were dipped in the melt and connected to the electrometer. The current ﬂowing between Polymers 2021, 13, x FOR PEER REVIEW 5 of 18 the electrodes was measured at 10 V (Figure2). The electrical resistance was measured 3 times at each temperature and the mean is presented for comparison.

Figure 2. Electrometer apparatus for the analysis of electrical resistance in the PBS melt. Figure 2. Electrometer apparatus for the analysis of electrical resistance in the PBS melt. 2.2.3. Characterization of PBS Fibers 2.2.3. FiberCharacterization diameters were of PBS measured Fibers using an Olympus BX53 microscope fitted with an OlympusFiber DP26diameters camera were (Olympus, measured Leiderdorp, using an TheOlympus Netherlands). BX53 microscope For each sample, fitted with the fiber an Olympusdiameter wasDP26 measured camera (Olympus, 100 times in Leiderdorp, different positions The Netherlands). based on the For 50× eachmagnified sample, image. the fiber diameterDifferential was scanning measured calorimetry 100 times (DSC)in different was carriedposition outs based using on the the Q2000 50× magnified device fo- image.cusing on changes to the glass transition temperature (Tg), melting temperature (Tm) and Differentialcrystallinity scanning (Xc) caused calorimetry by different (DSC) process was parameters.carried out using DSC was the appliedQ2000 device to PBS focusing granules as well as fibers with the smallest and largest diameters. Any change in fiber crystallinity on changes to the glass transition temperature (Tg), melting temperature (Tm) and during spinning is typically caused by the drawing process. Therefore, we selected samples crystallinity (Xc) caused by different process parameters. DSC was applied to PBS that had undergone the most drawing (smallest diameter) and the least drawing (biggest granules as well as fibers with the smallest and largest diameters. Any change in fiber crystallinity during spinning is typically caused by the drawing process. Therefore, we selected samples that had undergone the most drawing (smallest diameter) and the least drawing (biggest diameter) to compare with the PBS granules. No change in crystallinity indicated that the process parameter did not influence the physical properties of PBS. We used a temperature range of –40 to +300 °C and a heating rate of 10 °C/min. The melt enthalpy of 100% crystalline PBS was considered to be 110.3 J/g [39]. For each sample, we made 3 measurements and the mean values are presented. The effect of processing on the molecular weight of PBS was determined by gel per- meation chromatography (GPC) using a 1260 Infinity System (Agilent Technologies, Santa Clara, USA). We used hexafluor-2-isopropanol (HFIP) containing 0.19% sodium trifluoro- acetate as the mobile phase, flowing at a rate of 0.33 mL/min. GPC was used to compare PBS granules with fibers spun at the lowest throughput at different processing temperatures (235 and 265 °C). By testing the low-throughput samples (longest dwell time), we were able to conclude that no degradation under these conditions would infer the absence of degradation at higher throughputs. Solutions were prepared by dissolving 5-mg samples in HFIP for ~2 h before passing through a 0.2-µm polytetrafluoroethylene filter and injecting them into a modified silica column filled with 7-µm particles (Polymer Standards Service, Mainz, Germany). The relative molecular weight (Mw), number average molar mass (Mn), and polydispersity index (PDI) were determined using refractive index detectors calibrated with a standard polymethyl methacrylate polymer (1.0×105 g/mol). We performed GPC analysis 3 times with each sample and presented the mean values for comparison.

2.2.4. Statistical Analysis We used a full factorial design with four factors and two levels (Table 1) for multi- way analysis of variance (ANOVA) in SPSS (IBM, New York, NY, USA) to determine the statistical significance of each parameter with a general univariate analysis. We tested the effect of independent factors (temperature, electric field strength, throughput and the distance between nozzle and collector) and their interactions on the fiber diameter. We also compared effect strength η2 values of different process parameters to determine which exerted the greatest influence.

Polymers 2021, 13, 1024 5 of 16

diameter) to compare with the PBS granules. No change in crystallinity indicated that the process parameter did not influence the physical properties of PBS. We used a temperature range of −40 to +300 ◦C and a heating rate of 10 ◦C/min. The melt enthalpy of 100% crystalline PBS was considered to be 110.3 J/g [39]. For each sample, we made 3 measurements and the mean values are presented. The effect of processing on the molecular weight of PBS was determined by gel perme- ation chromatography (GPC) using a 1260 Infinity System (Agilent Technologies, Santa Clara, CA, USA). We used hexafluor-2-isopropanol (HFIP) containing 0.19% sodium trifluoroac- etate as the mobile phase, flowing at a rate of 0.33 mL/min. GPC was used to compare PBS granules with fibers spun at the lowest throughput at different processing temperatures (235 and 265 ◦C). By testing the low-throughput samples (longest dwell time), we were able to conclude that no degradation under these conditions would infer the absence of degradation at higher throughputs. Solutions were prepared by dissolving 5-mg samples in HFIP for ~2 h before passing through a 0.2-µm polytetrafluoroethylene filter and injecting them into a modified silica column filled with 7-µm particles (Polymer Standards Service, Mainz, Germany). The relative molecular weight (Mw), number average molar mass (Mn), and polydispersity index (PDI) were determined using refractive index detectors calibrated with a standard polymethyl methacrylate polymer (1.0 × 105 g/mol). We performed GPC analysis 3 times with each sample and presented the mean values for comparison.

2.2.4. Statistical Analysis We used a full factorial design with four factors and two levels (Table1) for multi- way analysis of variance (ANOVA) in SPSS (IBM, New York, NY, USA) to determine the statistical significance of each parameter with a general univariate analysis. We tested the effect of independent factors (temperature, electric field strength, throughput and the distance between nozzle and collector) and their interactions on the fiber diameter. We Polymers 2021, 13, x FOR PEER REVIEW 2 6 of 18 also compared effect strength η values of different process parameters to determine which exerted the greatest influence.

3. 3.Results Results and and Discussion Discussion 3.1.3.1. Thermal Thermal Analysis Analysis of the of Polymer the Polymer TGATGA revealed revealed that that PBS PBSgranules granules undergo undergo single-step single-step degradation degradation (Error! Reference (Figure3), with 5% sourceweight not loss found. at 347), with◦C and5% weight 50% weight loss at loss347 °C at 398and ◦50%C, in weight agreement loss at with 398 °C, previous in agree- reports [40]. mentThe with processing previous temperaturesreports [40]. The we processi selectedng temperatures are lower than we selected the degradation are lower than temperatures thedetermined degradation by temperatures TGA. determined by TGA.

FigureFigure 3. Thermogravimetric 3. Thermogravimetric analysis analysis (TGA) thermogram (TGA) thermogram showing the showing temperature-dependent the temperature-dependent degradationdegradation profile proﬁle of PBS of PBSgranules. granules.

3.2. Effect of Temperature on Viscosity The complex viscosity of PBS was plotted as a function of angular frequency at different temperatures (Error! Reference source not found.a). This revealed that the melts approach a plateau of Newtonian behavior at low angular frequencies, but non-Newto- nian (pseudoplastic) behavior is observed as the angular frequency increases. Accord- ingly, the complex viscosity declines sharply, as previously reported for PBS [41]. Entan- glements and chain interactions such as van der Waals forces and hydrogen bonds in the polymer can hinder the polymer flow. At lower shear, the polymer chains move so slowly that these interactions do not impede the flow and the shear is not sufficient to break these interactions. Therefore, we observed Newtonian behavior, where viscosity is independent of shear. Such interactions can be broken by increasing the shear and/or the temperature. As the shear rate increases and the interactions are broken, the polymer chains become oriented in the direction of shear making the polymer flow easier. The declining complex viscosity we observed with increasing angular frequency thus reflected the corresponding increase in shear. To visualize the effect of temperature on the complex viscosity, the complex viscosity of PBS as a function of temperature is shown in Error! Reference source not found.b at an angular frequency of 10 rad/s.

Polymers 2021, 13, 1024 6 of 16

3.2. Effect of Temperature on Viscosity The complex viscosity of PBS was plotted as a function of angular frequency at different temperatures (Figure4a). This revealed that the melts approach a plateau of Newtonian behavior at low angular frequencies, but non-Newtonian (pseudoplastic) behavior is observed as the angular frequency increases. Accordingly, the complex viscosity declines sharply, as previously reported for PBS [41]. Entanglements and chain interactions such as van der Waals forces and hydrogen bonds in the polymer can hinder the polymer flow. At lower shear, the polymer chains move so slowly that these interactions do not impede the flow and the shear is not sufficient to break these interactions. Therefore, we observed Newtonian behavior, where viscosity is independent of shear. Such interactions can be broken by increasing the shear and/or the temperature. As the shear rate increases and the interactions are broken, the polymer chains become oriented in the direction of shear making the polymer flow easier. The declining complex viscosity we observed with increas- Polymers 2021, 13, x FOR PEER REVIEW 7 of 18 ing angular frequency thus reflected the corresponding increase in shear. To visualize the effect of temperature on the complex viscosity, the complex viscosity of PBS as a function of temperature is shown in Figure4b at an angular frequency of 10 rad/s.

Figure 4. (a) RheogramRheogram showingshowing the the complex complex viscosity viscosity of of PBS PBS melts melts with with increasing increasing angular angular frequency fre- quencyat different at different temperatures; temperatures; (b) Complex (b) Complex viscosity viscosity of PBS asof aPBS function as a function of temperature of temperature at an angular at an angularfrequency frequency of 10 rad/s. of 10 Data rad/s. are Data means are withmeans standard with standard deviation deviation (n = 3). (n = 3).

As anticipated, the complex viscosity of PBS declined with increasing temperature. There areare twotwo potentialpotential explanations explanations for for this this observation. observation. First, First, as as set set out out above, above, the the entan- en- tanglementsglements and and chain chain interactions interactions arebroken are broken at high at high temperatures, temperatures, facilitating facilitating the flow the of flow the polymerof the polymer chains. chains. Similar Similar results results have been have reported been reported for PLA for [42 ].PLA We [42]. found We that found the complex that the viscositycomplex ofviscosity PBS declined of PBS by declined ~42% when by ~42% the temperaturewhen the temperature increased from increased 145 to from 175 ◦ C,145 and to 175by ~35% °C, and when by ~35% the temperature when the temperature increased from increased 175 to 205 from◦C. 175 However, to 205 °C. as weHowever, move further as we ◦ movefrom the further melting from point the (110meltingC), point there (110 is a less °C), significant there is a less decline significant in viscosity decline with in increasing viscosity withtemperatures. increasing The temperatures. decline in viscosity The decline was only in viscosity ~20% when was theonly temperature ~20% when increased the tempera- from ◦ ture205 to increased 235 C and from a similar 205 to 235 value °C was and observed a similar whenvalue thewas temperature observed when increased the temperature from 235 to ◦ increased265 C, perhaps from 235 because to 265 most °C, ofperhaps the chain becaus interactionse most of and the entanglementschain interactions in the and polymer entan- ◦ glementsare already in broken the polymer at 205 areC so already increasing broken the temperatureat 205 °C so hasincreasing no further the effect. temperature The second has potentialno further explanation effect. The issecond the degradation potential explan of theation polymer is the at higherdegradation temperatures, of the polymer which we at higherinvestigated temperatures, by GPC (Sectionwhich we 3.4 investigated). by GPC (Section 3.4). 3.3. Effect of Temperature on Electrical Conductivity 3.3. Effect of Temperature on Electrical Conductivity The electrical resistance of PBS was measured at melt temperatures of 145, 175, 205, The electrical resistance of PBS was measured at melt temperatures of 145, 175, 205, 235 and 265 ◦C to determine the effect of temperature on conductivity. We found that 235 and 265 °C to determine the effect of temperature on conductivity. We found that higher temperatures generally reduced the electrical resistance (Figure5). The resistance of higher temperatures generally reduced the electrical resistance (Error! Reference source the polymer melt was ~8 GΩ at 145 ◦C, but increasing the temperature by 30 ◦C to 175 ◦C not found.). The resistance of the polymer melt was ~8 GΩ at 145 °C, but increasing the reduced the electrical resistance by 10-fold. Further reductions, albeit much smaller in temperature by 30 °C to 175 °C reduced the electrical resistance by 10-fold. Further reduc- magnitude, were observed at 205, 235 and 265 ◦C. The lowest electrical resistance of 20 MΩ tions, albeit much smaller in magnitude, were observed at 205, 235 and 265 °C. The lowest was observed at 265 ◦C. electrical resistance of 20 MΩ was observed at 265 °C. Polymers are insulators and the electrical conductivity is intrinsically dependent on the temperature [43]. For polymers with polar functional groups (such as PBS), higher temperatures increase the mobility of the polymer chains, leading to ionic conduction via the polar groups and micro-Brownian motion [44]. The observed trend in electrical resistance as a function of temperature supports the rheological data (Error! Reference source not found.b). The change in viscosity of the polymer reached a plateau at temperatures exceeding 205 °C. Increasing the temperature beyond this point did not substantially increase the mobility of the polymer chains. The change in electrical resistance followed the same trend, further supporting the hypothesis that ionic conduction in PBS is mediated by the mobility of the polymer chains.

Polymers 2021, 13, x FOR PEER REVIEW 8 of 18 Polymers 2021, 13, 1024 7 of 16

Polymers 2021, 13, x FOR PEER REVIEW 8 of 18

FigureFigure 5. 5. TheThe electrical electrical resistance resistance (M (MΩΩ) of) of PBS PBS melts melts as as a function a function of of temperature. temperature. Data Data are are means means withwith ±± standardstandard deviation deviation (n ( n= =3). 3).

3.4. MolecularPolymers Weight are insulators of the Melt and Electrospun the electrical Fibers conductivity is intrinsically dependent on the temperature [43]. For polymers with polar functional groups (such as PBS), higher temperaturesThe GPC curves increase of theunprocessed mobility ofPBS the granules polymer and chains, PBS fibers leading processed to ionic at conduction 235 and 265via °C the are polar presented groups in and Error! micro-Brownian Reference source motion not [found.44]. The. observed trend in electrical resistance as a function of temperature supports the rheological data (Figure4b). The change in viscosity of the polymer reached a plateau at temperatures exceeding 205 ◦C. Increasing

the temperature beyond this point did not substantially increase the mobility of the polymer Figurechains. 5. The The electrical change resistance in electrical (MΩ resistance) of PBS melts followed as a function the same of temperature. trend, further Data supporting are meansthe with ± standard deviation (n = 3). hypothesis that ionic conduction in PBS is mediated by the mobility of the polymer chains.

3.4.3.4. Molecular Molecular Weight Weight of ofthe the Melt Melt Electrospun Electrospun Fibers Fibers TheThe GPC GPC curves curves of ofunprocessed unprocessed PBS PBS granules granules and and PBS PBS fibers ﬁbers processed processed at at235 235 and and 265265 °C◦ areC are presented presented in inError! Figure Reference6. source not found..

Figure 6. The molar mass distribution curve of unprocessed PBS granules (G) and PBS fibers processed at 235 and 265 °C.

The relative Mw, Mn and PDI of the same samples are presented in Error! Reference source not found..

FigureFigure 6. 6.TheThe molar molar mass mass distribution distribution curve curve of of unproc unprocessedessed PBS granulesgranules (G)(G) and and PBS PBS fibers fibers processed pro- cessedat 235 at and 235 265and◦ 265C. °C.

TheThe relative relative Mw, Mw, Mn Mn and and PDI PDI of ofthe the same same samples samples are are presented presented in inError! Figure Reference7. source not found..

Polymers 2021, 13, x FOR PEER REVIEW 9 of 18

Polymers 2021, 13, 1024 8 of 16

Figure 7. Weight average relative molecular weight (Mw), number average molar mass (Mn) and Figure 7. Weight average relative molecular weight (Mw), number average molar mass (Mn) and polydispersity index (PDI) of unprocessed PBS granules (G) and PBS fibers processed at 235 and polydispersity index (PDI) of unprocessed PBS granules (G) and PBS fibers processed at 235 and ◦ 265265 °C. DataC. Data are aremeans means with with standard standard deviation deviation (n = 3). (n = 3). The Mn, Mw and PDI of unprocessed PBS granules were 27,210 Da, 88,760 Da and 3.26, The Mn, Mw and PDI of unprocessed PBS granules were 27,210 Da, 88,760◦ Da and 3.26,respectively. respectively. The The Mn, Mn, Mw Mw and and PDI PDI ofof thethe PBS fibers fibers processed processed at at 235 235 °C Cwere were 26,210 26,210 Da, ◦ Da,77,350 77,350 Da Da and and 2.95, 2.95, respectively. respectively. The The corresponding corresponding values valuesfor for thethe fibersfibers processedprocessed atat 265 C 265were °C were 30,660 30,660 Da, 96,340Da, 96,340 Da and Da 3.14,and 3.14, respectively. respectively. The resultsThe results indicated indicated that processingthat pro- PBS cessingat these PBS specific at these temperatures specific temperatures resulted in re nosulted significant in no significant change in change the MW, in Mnthe orMW, PDI. This Mnconfirms or PDI. This that anyconfirms reduction that any in viscosityreduction is in explained viscosity byis explained the higher by temperature the higher tem- increasing peraturechain mobilityincreasing and chain not mobility by polymer and not degradation. by polymer degradation.

3.5.3.5. Fiber Fiber Diameter Diameter and andDistribution Distribution WeWe next next investigated investigated the theprocessability processability of PB ofS and PBS the and effect the effectof temperature, of temperature, electric electric field,field, throughput throughput and and nozzle-to-collector nozzle-to-collector dist distanceance on onthe thefiber fiber diameter. diameter. When When the thepoly- polymer mermelt melt droplet droplet becomes becomes charged charged in in a a field field ofof sufficientsufficient strength,strength, the electrostatic electrostatic repul- repulsion is sionstrong is strong enough enough to overcome to overcome the the surface surfac tensione tension and and stretch stretch the the droplet. droplet. When When this this charge chargebecomes becomes higher higher than than a certain a certain threshold, threshold, a a jet jet knownknown as a a Taylor Taylor cone cone erupts erupts from from the thepolymer polymer dropletdroplet [45]. [45]. We We observed observed the the formation formation of a of Taylor a Taylor cone cone under under all the all process the process conditionsconditions we wetested. tested. Error!Figure Reference8 shows source the average not found. fiber shows diameter the average ( ±standard fiber diameter deviation) (±standard produced de- at spin- viation)neret temperaturesproduced at spinneret of 235 and temperatures 265 ◦C, with of 235 nozzle- and 265 to-collector °C, with distancesnozzle- to-collector of 8 and 10 cm, distancesthroughputs of 8 and of 10 0.1 cm, and throughputs 4 mL/min andof 0.1 the an electricd 4 mL/min field and strength the electric set to 50field kV. strength Reducing the setthroughput to 50 kV. Reducing from 4 tothe 0.1 throughput mL/min generatedfrom 4 to 0.1 finer mL/min fibers generated at both temperatures. finer fibers at Because both the temperatures.polymer melt Because emerging the polymer from the melt nozzle emerging is pulled from down the bynozzle gravity is pulled and by down its interaction by gravitywith theand electricby its interaction field, we hypothesized with the electr thatic field, reducing we hypothesized the throughput that givesreducing more the time for throughputthe polymer gives to more interact time with for the the polymer electric field,to interact increasing with the the electric probability field, increasing of whipping be- thehavior. probability The fiberof whipping diameter be producedhavior. The at fiber a throughput diameter prod of 0.1uced mL/min at a throughput was also significantlyof 0.1 mL/minreduced was by also increasing significantly the reduced nozzle-to-collector by increasing distance the nozzle-to-collector from 8 to 10 cm. distance For example,from at 8 to265 10 ◦cm.C and For aexample, throughput at 265 of °C 0.1 and mL/min, a throughput increasing of 0.1 themL/min, distance increasing from 8 the to 10distance cm reduced fromthe 8 average to 10 cm fiber reduced diameter the average by 15.6% fiber (from diameter 65 to 54by µ15.6%m). As (from proposed 65 to 54 above, µm). increasingAs pro- the posednozzle-to-collector above, increasing distance the nozzle-to-collector would also give thedistance fiber morewould time also to give interact the fiber with more the electric timefield, to interact thereby with promoting the electric whipping field, th behavior.ereby promoting But if the whipping collector behavior. were placed But tooif the far from collectorthe nozzle, were thereplaced would too far be from less interactionthe nozzle, withthere thewould electric be less field interaction leading to with thicker the fibers. electricUnder field similar leading conditions to thicker with fibers. a throughput Under similar of 4conditions mL/min, with the same a throughput change inof distance4 mL/min, the same change in distance achieved only a 6.7% reduction in fiber diameter. A achieved only a 6.7% reduction in fiber diameter. A similar trend was observed at 235 ◦C. similar trend was observed at 235 °C. When increasing the throughout from 0.1 to 4 When increasing the throughout from 0.1 to 4 mL/min, the material flow increased 40-fold. Given this higher material flow, it is likely that increasing the nozzle-to-collector distance by 2 cm is not sufficient to provide enough additional time for the fibers to interact with the field. Polymers 2021, 13, x FOR PEER REVIEW 10 of 18

mL/min, the material flow increased 40-fold. Given this higher material flow, it is likely Polymers 2021, 13, 1024 9 of 16 that increasing the nozzle-to-collector distance by 2 cm is not sufficient to provide enough additional time for the fibers to interact with the field.

FigureFigure 8. Diameter 8. Diameter of PBS of PBSfibers ﬁbers at polymer at polymer melt temp melteratures temperatures of 235 ofor 235265 or°C, 265 throughputs◦C, throughputs of 0.1 of 0.1 or or 4 mL/min, a anozzle-to-collector nozzle-to-collector distance distance of 8 of or 8 10 or cm, 10 cm, and and the electric the electric fieldﬁeld strength strength set to set 50 to 50 kV. The kV.x The-axis x-axis shows shows the temperaturethe temperature with with the the nozzle-to-collector nozzle-to-collectordistance distance in parentheses. parentheses. Data Data are means are means with standard deviations (n = 100). with standard deviations (n = 100).

WeWe found found that that the temperature the temperature had no had signif noicant significant effect on effect fiber on diameter fiber diameter compared compared to tothroughput throughput or nozzle-to-collector or nozzle-to-collector distance distance.. For example, For example, the mean the fiber mean diameter fiber diameter at a at a throughput of 0.1 mL/min and a nozzle-to-collector distance of 10 cm was 56 µm at 235 throughput of 0.1 mL/min and a nozzle-to-collector distance of 10 cm was 56 µm at 235 ◦C °C and 54.22 µm at 265 °C, the latter being the narrowest fiber produced at 50 kV. How- and 54.22 µm at 265 ◦C, the latter being the narrowest fiber produced at 50 kV. However, ever, much finer fibers were produced when we increased the electric field strength to 60 much finer fibers were produced when we increased the electric field strength to 60 kV kV (Error! Reference source not found.). Although Taylor-cone formation and whipping behavior(Figure occurred9). Although under Taylor-cone all the process formation parameters and we whipping tested, the behavior whipping occurred behavior under was all the enhancedprocess at parameters 60 kV. Accordingly, we tested, the the average whipping fiber behavior diameter wasat a enhancedtemperature at of 60 235 kV. °C, Accordingly, a ◦ throughputthe average of 0.1 fiber mL/min, diameter and at a anozzle-to-collector temperature of 235 distanceC, a throughputof 10 cm was of reduced 0.1 mL/min, by and a Polymers 2021, 13, x FOR PEER REVIEW 11 of 18 46%nozzle-to-collector (from 56 to 30 µm) distance in the stronger of 10 cm electric was reducedfield. This by result 46% is (from consistent 56 to 30withµm) previous in the stronger studieselectric [2]. field. This result is consistent with previous studies [2].

Figure 9. 9. DiameterDiameter of of PBS PBS fibers fibers at atpolymer polymer melt melt temp temperatureseratures of 235 of 235or 265 or 265°C, throughputs◦C, throughputs of 0.1 of 0.1 or or4 mL/min, 4 mL/min, a a nozzle-to-collector nozzle-to-collector distancedistance of 8 or 10 10 cm, cm, and and the the electric electric field field strength strength set set to to60 60 kV. The kV. The x-axis shows the temperature with the nozzle-to-collector distance in parentheses. Data x-axis shows the temperature with the nozzle-to-collector distance in parentheses. Data are means are means with standard deviations (n = 100). with standard deviations (n = 100). The trend toward narrower fibers at a lower process throughput was maintained at The trend toward narrower fibers at a lower process throughput was maintained at the higher voltage. By reducing the throughput from 4 to 0.1 mL/min, the average fiber the higher voltage. By reducing the throughput from 4 to 0.1 mL/min, the average fiber diameter decreased by 30% (from 50 to 35 µm) at a temperature of 235 °C and a nozzle-to collector distance of 8 cm. This phenomenon was observed regardless of the processing temperature or nozzle-to-collector distance. When we compared fiber diameters as a function of temperature while keeping the remaining parameters constant, we found that temperature had no significant influence. An interesting phenomenon observed during high-throughput processing at 265 °C and 60 kV was that changing the nozzle-to-collector distance did not affect the fiber diameter. The average fiber diameter at a throughput of 4 mL/min was 33 and 32 µm when the nozzle-to-collector distance was 8 and 10 cm, respectively. The viscosity and resistance of PBS were also marginally lower at 265 than at 235 °C. As the material comes through the nozzle, fibers are formed due to a combination of gravitational pull and the electric field. One hypothesis to explain the effect observed at the high temperature in the strongest electric field is that the material flows so quickly under these conditions that increasing the nozzle to collector distance by 2 cm does not increase the whipping behavior and thus has no significant effect on the fiber diameter. When the electric field strength was 60 kV, the finest fibers (30.05 µm) were produced at 235 °C with a throughput of 0.1 mL/min and a nozzle-to-collector distance of 10 cm. The Taylor cone formation observed during melt electrospinning and microscopy images of the resulting PBS fibers produced are presented in Error! Reference source not found..

Polymers 2021, 13, 1024 10 of 16

diameter decreased by 30% (from 50 to 35 µm) at a temperature of 235 ◦C and a nozzle-to collector distance of 8 cm. This phenomenon was observed regardless of the processing temperature or nozzle-to-collector distance. When we compared fiber diameters as a function of temperature while keeping the remaining parameters constant, we found that temperature had no significant influence. An interesting phenomenon observed during high-throughput processing at 265 ◦C and 60 kV was that changing the nozzle-to-collector distance did not affect the fiber diameter. The average fiber diameter at a throughput of 4 mL/min was 33 and 32 µm when the nozzle-to-collector distance was 8 and 10 cm, respectively. The viscosity and resistance of PBS were also marginally lower at 265 than at 235 ◦C. As the material comes through the nozzle, fibers are formed due to a combination of gravitational pull and the electric field. One hypothesis to explain the effect observed at the high temperature in the strongest electric field is that the material flows so quickly under these conditions that increasing the nozzle to collector distance by 2 cm does not increase the whipping behavior and thus has no significant effect on the fiber diameter. When the electric field strength was 60 kV, the finest fibers (30.05 µm) were produced at 235 ◦C with a throughput of 0.1 mL/min and a nozzle-to-collector distance of 10 cm. The Taylor cone formation observed during melt Polymers 2021, 13, x FOR PEER REVIEWelectrospinning and microscopy images of the resulting PBS fibers produced are12 of presented 18

in Figure 10.

FigureFigure 10. 10.TaylorTaylor cone cone formation formation during during melt meltelectros electrospinningpinning and images and images of the ofresulting the resulting fibers. fibers. (a) (a)Taylor Taylor cone coneformation. formation. ((bb–d) Microscopy images of of PBS PBS electr electrospunospun fibers fibers under under different different pro- processing cessing conditions. (b) Temperature: 235 °C, throughput: 0.1 mL/min, electric field: 60 kV, nozzle- conditions. (b) Temperature: 235 ◦C, throughput: 0.1 mL/min, electric field: 60 kV, nozzle-to-collector to-collector distance: 10 cm. (c) Temperature:◦ 265 °C, throughput: 0.1 mL/min, electric field: 60 kV, nozzle-to-collectordistance: 10 cm. distance (c) Temperature: 8 cm. (d) Temperature: 265 C, throughput: 265 °C, throughput 0.1 mL/min,: 4 mL/min, electric electric field: field: 60 kV, 50 nozzle- ◦ kV,to-collector nozzle-to-collector distance distance: 8 cm. ( d8) cm. Temperature: 265 C, throughput: 4 mL/min, electric field: 50 kV, nozzle-to-collector distance: 8 cm. The melt electrospinning of PLA under similar conditions (identical apparatus, electric fieldThe strength melt electrospinning 60 kV, nozzle-to-collector of PLA under distance similar 10 conditions cm, throughput (identical 4 mL/min, apparatus, tem- electric peraturefield strength 300 °C) 60generated kV, nozzle-to-collector fibers with an av distanceerage diameter 10 cm, throughput of 112.5 µm, 4 mL/min,whereas at tempera- a ◦ lowerture temperature 300 C) generated of 235 °C, fibers the with diameter an average of PBS diameterfiber produced of 112.5 wereµm, 43.42 whereas µm (61% at a lower ◦ lower).temperature These results of 235 suggestC, the that diameter PBS is more of PBS suitable fiber producedas a material were for 43.42melt electrospin-µm (61% lower). ning than PLA [34]. Statistical analysis of the datasets in Error! Reference source not found. andError! Reference source not found. by multi-way ANOVA revealed the statistical significance of differences in fiber diameter as a function of the process parameters (temperature, throughput, electric field strength, and nozzle-to-collector distance) and their interactions. The relationships between individual process parameters (factors) and the average fiber diameter are summarized in Error! Reference source not found.. Narrower fibers were produced by reducing the process throughput or by increasing the electric field strength or nozzle-to-collector distance. Higher temperatures also tended to produce narrower fibers but the effect of this parameter was not significant.

Polymers 2021, 13, 1024 11 of 16

These results suggest that PBS is more suitable as a material for melt electrospinning than PLA [34]. Statistical analysis of the datasets in Figures8 and9 by multi-way ANOVA revealed the statistical significance of differences in fiber diameter as a function of the process parameters (temperature, throughput, electric field strength, and nozzle-to-collector distance) and their interactions. The relationships between individual process parameters (factors) and the average fiber diameter are summarized in Figure 11. Narrower fibers were produced by reducing the process throughput or by increasing the electric field strength or nozzle-to- Polymers 2021, 13, x FOR PEER REVIEW 13 of 18 Polymers 2021, 13, x FOR PEER REVIEW 13 of 18 collector distance. Higher temperatures also tended to produce narrower fibers but the effect of this parameter was not significant.

FigureFigure 11. 11. VariationsVariations in in PBSPBS PBS fiberfiber ﬁber diameterdiameter diameter as as a asa fu fu anctionnction function of of different different of different process process process parameters. parameters. parameters. (a ()a Elec-) Elec- (a) Electric trictricﬁeld field strength strength vs. vs.vs. diameter, diameter,diameter, (b ()(bb throughput)) throughputthroughput vs. vs. vs. diameter, diameter, diameter, ((c c())c nozzle-to-collector)nozzle-to-collector nozzle-to-collector distance distancedistance vs. vs.vs. diameter, diameter, (d) temperature vs. diameter. diameter,(d) temperature (d) temperature vs. diameter. vs. diameter.

TheThe effective effective strength strength ofof of thethe the variousvarious various factors factors factors we we considered weconsidered considered and and their and their theirinteractions interactions interactions are are are summarized below in Error! Reference source not found.. summarizedsummarized below below in in Error! Figure Reference 12. source not found..

Figure 12. Pareto chart representing the effective strength of the factors we tested and their inter- Figureactions.Figure 12. 12.EF Pareto =Pareto electric chart chart field, representing representing Th = throughput, the the effective effectiveD = nozzle-to-collector streng strengthth of the of thefactors distance, factors we testedTe we = testedtemperature. and their and theirinter- interac- actions. EF = electric field, Th = throughput, D = nozzle-to-collector distance, Te = temperature. tions. EF = electric ﬁeld, Th = throughput, D = nozzle-to-collector distance, Te = temperature. The significance level of the factors and their interactions are presented in Error! Ref- erenceThe source significance not found. level. of the factors and their interactions are presented in Error! Ref- erence source not found..

Polymers 2021, 13, 1024 12 of 16

The signiﬁcance level of the factors and their interactions are presented in Table2.

Table 2. Signiﬁcance level of the factors we tested and their interactions. EF = electric ﬁeld, Th = throughput, D = nozzle-to-collector distance, Te = temperature.

Factor Signiﬁcance Level (p) EF <0.001 Th <0.001 D <0.001 Te 0.209 Te*EF <0.001 Te*Th 0.015 Te*D 0.113 EF*Th 0.001 EF*D 0.009 Th*D 0.183 Te*EF*Th 0.003 Te*EF*D 0.055 Te*Th*D 0.439 EF*Th*D 0.051 Te*EF*Th*D 0.343

Statistical analysis revealed that the electric field, throughput, and nozzle-to-collector distance all had a significant impact (p < 0.05) on the diameter of the fiber whereas temperature was not statistically significant (p > 0.05). This agrees with the data reported earlier. Under fixed conditions of a 0.1 mL/min throughput, a 10 cm nozzle-to-collector distance and a 50 kV electric field, the diameter of the fiber obtained only changed from 56 µm at 235 ◦C to 54.22 µm at 265 ◦C. The possible hypothesis for this was explained using the changes observed in the rheological behavior and electrical conductivity of the polymer as a function of temperature. We also observed statistically significant interactions (p < 0.05) between the factors temperature and electric field, temperature and throughput, electric field and throughput, electric field and nozzle-to-collector distance, and finally temperature, electric field and throughput. As shown in Figure 12, the electric field has the strongest influence on the fiber diameter, followed by throughput, then nozzle-to-collector distance. This also agrees with the data presented earlier. Among all the factors we tested, increasing the electric field from 50 kV to 60 kV led to the most substantial reduction in fiber diameter.

3.6. Thermal Properties of the Electrospun Fibers The DSC thermograms of PBS granules and PBS fibers with the highest diameter (76.68 µm) and lowest diameter (30.05 µm) produced under our processing conditions are presented in Figure 13. The Tg, Tm and Xc values are summarized in Table3. The DSC thermograms of the granules and both fibers featured a melting peak (Tm2) at ~110 ◦C, which is also the PBS melting point reported by the manufacturer. The values we observed agree with those reported in previous studies [46]. Another small peak was observed at ~50 ◦C, possibly representing an additive such as a plasticizer used to improve the processability of PBS. The PBS granules feature a second melting peak just below 100 ◦C (Tm1). This is mirrored by exothermic peaks in the thermograms for each of the fibers (Trc) possibly associated with the recrystallization (or recrystallization and melting) of PBS. The crystallization process occurring during fiber formation can involve the formation of crystals differing in size and containing various defects. As the PBS is heated, these small crystals or in some cases, crystal defects, can melt and recrystallize or combine to form larger crystals, which melt again at the higher temperature. This could also be from amorphous polymer chains, due to higher mobility at these temperatures, forming a structure that is more ordered and therefore leading to crystallization. Similar observations Polymers 2021, 13, 1024 13 of 16

have been reported in earlier studies [47]. The enthalpies of melting of the melting peaks at 35 ◦C were not considered for the calculation of Xc because they are thought to correspond to the melting of an additive. However, the enthalpy of recrystallization was taken into Polymers 2021, 13, x FOR PEER REVIEW 15 of 18 account and subtracted from the melting enthalpy to calculate the total crystallinity of PBS. In the case of PBS granules, the enthalpy of melting from the ﬁrst melting peak (Tm1) was taken into consideration and added to the enthalpy obtained from the Tm2 to obtain the Xc value.The The DSC Xc thermograms of all samples of remained PBS granules constant and at PBS ~57%. fibers Our with DSC the results highest therefore diameter suggest (76.68that the µm) range and lowest of process diameter parameters (30.05 µm) we testedproduced did under not affect our processing the physical conditions properties are of the presentedPBS ﬁbers. in Error! Reference source not found..

FigureFigure 13. 13. DSCDSC thermogram thermogram of ofPBS PBS granules granules (G) (G) and and PBS PBS fibers ﬁbers with with the thesmallest smallest (S) and (S) andbiggest biggest (B) (B)diameters diameters we we achieved achieved using using our our processing processing conditions.conditions.

TableThe 3. TheTg, meltingTm and temperatureXc values are (Tm), summarized re-cooling in temperature Error! Reference (Trc) and source degree not of crystallinityfound.. (Xc) of PBS granules (G) and PBS fibers with the smallest (S) and biggest (B) diameters we achieved using Tableour processing 3. The melting conditions. temperature Data (Tm), are means re-cooling with temperature± standard deviations(Trc) and degree (n = 3). of crystallinity (Xc) of PBS granules (G) and PBS fibers with the smallest (S) and biggest (B) diameters we ◦ ◦ ◦ achievedMaterial using our processing Tm 1co(nditions.C) Data are Trc means ( C) with ± standard Tm2 ( deviationsC) (n = 3). Xc (%) G 91.1 ± 2 - 116.3 ± 2 59.97 ± 1 MaterialS Tm1 ( °C) - Trc 88.6( °C)± 3 Tm2 114.2( °C) ± 1 Xc (%) 55.6 ± 1 B - 90.4 ± 2 115.2 ± 2 58.9 ± 1 G 91.1 ± 2 - 116.3 ± 2 59.97 ± 1 4. Conclusions S - 88.6 ± 3 114.2 ± 1 55.6 ± 1 We have investigated the melt electrospinning of PBS microfibers using a single-nozzle laboratory-scaleB device, - testing different 90.4 process± 2 parameters 115.2 ± 2 (temperature, 58.9 ± electric 1 field strength, throughput and nozzle-to-collector distance) to determine their effect on fiber diameter. To the best of our knowledge, we have reported the first melt-electrospun PBS The DSC thermograms of the granules and both fibers featured a melting peak (Tm2) fibers produced in the low micrometer range. We observed the formation of a Taylor cone at ~110 °C, which is also the PBS melting point reported by the manufacturer. The values wefollowed observed by agree continuous with those fiber reported deposition in previous throughout studies the [46]. range Another of process small parameters peak was we µ ◦ observedtested. The at ~50 coarsest °C, possibly fibers representing (diameter = 74.05an additivem) were such producedas a plasticizer at 265 usedC to with improve an electric thefield processability strength of of 50 PBS. kV, aThe throughput PBS granules of 4feature mL/min a second and a melting nozzle-to-collector peak just below distance 100 of 8 cm. The whipping behavior was enhanced by increasing the electric field strength from °C (Tm1). This is mirrored by exothermic peaks in the thermograms for each of the fibers ◦ (Trc)50 kV possibly to 60 kV. associated The finest with fibers the (diameter recrystallization = 30.05 (orµm) recrystallization were produced and at 235 melting)C and of 60 kV, PBS.with The a throughput crystallization of 0.1 process mL/min occurring and a nozzle-to-collector during fiber formation distance can ofinvolve 10 cm. the for- mationWe of foundcrystals that differing low-throughput in size and meltcontaining electrospinning various defects. (0.1 mL/min) As the PBS reduced is heated, the fiber thesediameter smallat crystals both temperaturesor in some cases, and crystal field defects, strengths can we melt tested. and recrystallize The effect ofor changingcombine the tonozzle-to-collector form larger crystals, distance which from melt 8 again to 10 cmat the was higher also more temperature. significant This at lowcould throughput. also be It fromis likely amorphous that increasing polymer the chains, nozzle-to-collector due to higher mobility distance at during these temperatures, a low-throughput forming process aallows structure more that time is more for whipping ordered and behavior, therefore thus leading stretching to crystallization. the melt and Similar generating observa- narrower tionsfibers. have In contrast,been reported the fast in flowearlier of studies the polymer [47]. The during enthalpies the high-throughput of melting of the process melting means peaks at 35 °C were not considered for the calculation of Xc because they are thought to correspond to the melting of an additive. However, the enthalpy of recrystallization was

Polymers 2021, 13, 1024 14 of 16

that a 2-cm increase in the nozzle-to-collector distance does not have a significant effect on whipping behavior and thus on the fiber diameter. Multi-way ANOVA revealed that three factors (electric field strength, throughput, and nozzle-to-collector distance) had a significant effect on fiber diameter, whereas the effect of temperature was nonsignificant when the other factors remained constant. Our rheological and conductivity data showed that the change in viscosity and electrical conductivity was not significant within the temperature range we considered, and this may explain the results of the ANOVA test. Based on the effect strength η2, the electric field strength was the factor with the greatest influence on the fiber diameter. By varying four process parameters, we reduced the average diameter of our melt electrospun PBS fibers by 59% (from 74.05 to 30.05 µm). The narrowest fibers we produced were significantly finer than those we previously generated from PLA using the same experimental setup. Our results suggest that PBS could replace PLA and other polymers for the melt electrospinning of biomedical fibers. This provides a sustainable alternative to solution electrospinning, which is normally used to manufacture fibers for biomedical applications.

Author Contributions: Conceptualization, M.-E.O. and N.K.B.; methodology, M.-E.O. and N.K.B.; software, M.-E.O. and N.K.B.; validation, M.-E.O. and N.K.B.; formal analysis, M.-E.O. and N.K.B.; investigation, M.-E.O. and N.K.B.; resources, G.S.; data curation, M.-E.O. and N.K.B.; writing— original draft preparation, M.-E.O. and N.B; writing—review and editing, G.S. and R.G.; visualization, M.-E.O. and N.K.B.; supervision, G.S. and R.G.; project administration, M.-E.O. and N.K.B. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Data Availability Statement: The datasets used and/or analyzed during this study are available from the corresponding author on reasonable request. Acknowledgments: We thank Henri Becker for equipment maintenance and modification. We would also like to thank Kylie Koenig and Stefan Hermanns for helpful discussions. We also thank Julian Schmeling (MCPP, Germany) for donating the PBS granules. Conflicts of Interest: The authors declare no conflict of interest.

References 1. Manavitehrani, I.; Fathi, A.; Badr, H.; Daly, S.; Negahi Shirazi, A.; Dehghani, F. Biomedical Applications of Biodegradable Polyesters. Polymers 2016, 8, 20. [CrossRef] 2. Doustgani, A.; Ahmadi, E. Melt electrospinning process optimization of polylactic acid nanofibers. J. Ind. Text. 2016, 45, 626–634. [CrossRef] 3. Wu, F.; Huang, C.-L.; Zeng, J.-B.; Li, S.-L.; Wang, Y.-Z. Synthesis and characterization of segmented poly(butylene succinate) urethane ionenes containing secondary amine cation. Polymer 2014, 55, 4358–4368. [CrossRef] 4. Gigli, M.; Fabbri, M.; Lotti, N.; Gamberini, R.; Rimini, B.; Munari, A. Poly(butylene succinate)-based polyesters for biomedical applications: A review. Eur. Polym. J. 2016, 75, 431–460. [CrossRef] 5. Costa-Pinto, A.R.; Correlo, V.M.; Sol, P.C.; Bhattacharya, M.; Srouji, S.; Livne, E.; Reis, R.L.; Neves, N.M. Chitosan-poly(butylene succinate) scaffolds and human bone marrow stromal cells induce bone repair in a mouse calvaria model. J. Tissue Eng. Regen. Med. 2012, 6, 21–28. [CrossRef][PubMed] 6. Costa-Pinto, A.R.; Martins, A.M.; Castelhano-Carlos, M.J.; Correlo, V.M.; Sol, P.C.; Longatto-Filho, A.; Battacharya, M.; Reis, R.L.; Neves, N.M. In vitro degradation and in vivo biocompatibility of chitosan–poly(butylene succinate) fiber mesh scaffolds. J. Bioact. Compat. Polym. 2014, 29, 137–151. [CrossRef] 7. Stoyanova, N.; Paneva, D.; Mincheva, R.; Toncheva, A.; Manolova, N.; Dubois, P.; Rashkov, I. Poly(L-lactide) and poly(butylene succinate) immiscible blends: From electrospinning to biologically active materials. Mater. Sci. Eng. C Mater. Biol. Appl. 2014, 41, 119–126. [CrossRef][PubMed] 8. Lee, S.; Kay Obendorf, S. Developing protective textile materials as barriers to liquid penetration using melt-electrospinning. J. Appl. Polym. Sci. 2006, 102, 3430–3437. [CrossRef] 9. Shin, Y.M.; Hohman, M.M.; Brenner, M.P.; Rutledge, G.C. Electrospinning: A whipping fluid jet generates submicron polymer fibers. Appl. Phys. Lett. 2001, 78, 1149–1151. [CrossRef] 10. Bhat, G.S. Advances in polymeric nanofiber manufacturing technologies. J. Nanomater. Mol. Nanotechnol. 2016, 5, 1–2. [CrossRef] 11. Koenig, K.; Langensiepen, F.; Seide, G.; Daenicke, J.; Schubert, D. From Lab to Pilot Scale: Melt Electrospun Nanofibers of Polypropylene with Conductive Additives. J. Nanomater. Mol. Nanotechnol. 2018, 7, 1–8. [CrossRef] Polymers 2021, 13, 1024 15 of 16

12. Brown, T.; Dalton, P.; Hutmacher, D.W. Melt electrospinning today: An opportune time for an emerging polymer process. Prog. Polym. Sci. 2016, 56, 116–166. [CrossRef] 13. Sundarrajan, S.; Tan, K.L.; Lim, S.H. Electrospun nanofibers for air filtration applications. Procedia Eng. 2014, 75, 159–163. [CrossRef] 14. Wang, J.; Li, Y.; Sun, X. Challenges and opportunities of nanostructured materials for aprotic rechargeable lithium–air batteries. Nano Energy 2013, 2, 1459–1478. [CrossRef] 15. Wang, X.; Drew, C.; Lee, S.-H. Electrospun Nanofibrous membranes for highly sensitive optical sensors. Nano Lett. 2002, 2, 1273–1275. [CrossRef] 16. Research, G.V. Nanofibers Market Size, Share & Trends Analysis Report By Product (Polymer, Carbon, Composite, Cellulose, Metallic), By Application (Electronics, MCE, Energy, MLP), By Region, And Segment Forecasts, 2017–2024. Available online: https://www.grandviewresearch.com/industry-analysis/nanofibers-market (accessed on 18 March 2021). 17. Bhardwaj, N.; Kundu, S.C. Electrospinning: A fascinating fiber fabrication technique. Biotechnol. Adv. 2010, 28, 325–347. [CrossRef] [PubMed] 18. Ekram, B.; Abd El-Hady, B.M.; El-Kady, A.M.; Amr, S.M.; Gabr, H.; Waly, A.I.; Guirguis, O.W. Enhancing the Stability, Hy- drophilicity, Mechanical and Biological Properties of Electrospun Polycaprolactone in Formic Acid/Acetic Acid Solvent System. Fibers Polym. 2019, 20, 715–724. [CrossRef] 19. Van der Schueren, L.; Steyaert, I.; De Schoenmaker, B.; De Clerck, K. Polycaprolactone/chitosan blend nanofibres electrospun from an acetic acid/formic acid solvent system. Carbohydr. Polym. 2012, 88, 1221–1226. [CrossRef] 20. Koenig, K.; Balakrishnan, N.; Hermanns, S.; Langensiepen, F.; Seide, G. Biobased Dyes as Conductive Additives to Reduce the Diameter of Polylactic Acid Fibers during Melt Electrospinning. Materials 2020, 13, 1055. [CrossRef][PubMed] 21. Qin, Y.; Cheng, L.; Zhang, Y.; Chen, X.; Wang, X.; He, X.; Yang, W.; An, Y.; Li, H. Efficient preparation of poly(lactic acid) nanofibers by melt differential electrospinning with addition of acetyl tributyl citrate. J. Appl. Polym. Sci. 2018, 135, 46554. [CrossRef] 22. Ogata, N.; Shimada, N.; Yamaguchi, S.; Nakane, K.; Ogihara, T. Melt-electrospinning of poly(ethylene terephthalate) and polyalirate. J. Appl. Polym. Sci. 2007, 105, 1127–1132. [CrossRef] 23. Zaiss, S.; Brown, T.D.; Reichert, J.C.; Berner, A. Poly(epsilon-caprolactone) Scaffolds Fabricated by Melt Electrospinning for Bone Tissue Engineering. Materials 2016, 9, 232. [CrossRef][PubMed] 24. Jing, X.; Mi, H.Y.; Salick, M.R.; Cordie, T.M.; Peng, X.F.; Turng, L.S. Electrospinning thermoplastic polyurethane/graphene oxide scaffolds for small diameter vascular graft applications. Mater. Sci. Eng. C Mater. Biol. Appl. 2015, 49, 40–50. [CrossRef][PubMed] 25. Ko, J.; Kan, D.; Jun, M.B.G. Combining melt electrospinning and particulate leaching for fabrication of porous microfibers. Manuf. Lett. 2015, 3, 5–8. [CrossRef] 26. Hacker, C.; Seide, G.; Gries, T.; Thomas, H.; Moeller, M. Electrospinning of polymer melt: Steps toward an upscaled multi-jet process. In Proceedings of the International Conference Latest Advances in High Tech Textiles and Textile-Based Materials, Ghent, Belgium, 23–25 September 2009; pp. 71–76. 27. Phiriyawirut, M.; Sarapat, K.; Sirima, S.; Prasertchol, A. Porous Electrospun Nanofiber from Biomass-Based Polyester Blends of Polylactic Acid and Polybutylene Succinate. Open J. Polym. Chem. 2019, 9, 1–15. [CrossRef] 28. Cooper, C.J.; Mohanty, A.K.; Misra, M. Electrospinning Process and Structure Relationship of Biobased Poly(butylene succinate) for Nanoporous Fibers. ACS Omega 2018, 3, 5547–5557. [CrossRef] 29. Adomaviˇciut¯ e,˙ E.; Pupkeviˇciut¯ e,˙ S.; Juškaite,˙ V.; Žilius, M.; Stanys, S.; Pavilonis, A.; Briedis, V. Formation and Investigation of Electrospun PLA Materials with Propolis Extracts and Silver Nanoparticles for Biomedical Applications. J. Nanomater. 2017, 2017, 1–11. [CrossRef] 30. Koenig, K.; Hermanns, S.; Ellerkmann, J.; Saralidze, K.; Langensiepen, F.; Seide, G. The effect of additives and process parameters on the pilot-scale manufacturing of polylactic acid sub-microfibers by melt electrospinning. Text. Res. J. 1177, 1–14. [CrossRef] 31. Zhang, X.; Yang, X.; Wu, H.; Wang, H.; Su, Y.; Mi, H. Study on Performance and Structure of Electrospun Polybutylenes Succinate (PBS)/Meltblown Composite Materials. J. Inst. Eng. (India) Ser. E 2020, 1–5. [CrossRef] 32. Zhmayev, E.; Cho, D.; Joo, Y.L. Nanofibers from gas-assisted polymer melt electrospinning. Polymer 2010, 51, 4140–4144. [CrossRef] 33. Duncan Phillips, J.S.; Wilding, M.; Farrington, D.; Sandukas, S.; Boned, J.; Dervand, S. Effect of heat setting on dimensional stability and dyeing properties of poly(lactic acid) fibres. Coloration Technol. 2003, 119, 128–133. [CrossRef] 34. Balakrishnan, N.K.; Koenig, K.; Seide, G. The Effect of Dye and Pigment Concentrations on the Diameter of Melt-Electrospun Polylactic Acid Fibers. Polymers 2020, 12, 2321. [CrossRef][PubMed] 35. Takayama, T.; Todo, M. Improvement of impact fracture properties of PLA/PCL polymer blend due to LTI addition. J. Mater. Sci. 2006, 41, 4989–4992. [CrossRef] 36. Fabbri, M.; Soccio, M.; Costa, M.; Lotti, N.; Gazzano, M.; Siracusa, V.; Gamberini, R.; Rimini, B.; Munari, A.; García-Fernández, L.; et al. New fully bio-based PLLA triblock copoly(ester urethane)s as potential candidates for soft tissue engineering. Polym. Degrad. Stab. 2016, 132, 169–180. [CrossRef] 37. Vannaladsaysy, V.; Todo, M.; Takayama, T.; Jaafar, M.; Ahmad, Z.; Pasomsouk, K. Effects of lysine triisocyanate on the mode I fracture behavior of polymer blend of poly (l-lactic acid) and poly (butylene succinate-co-l-lactate). J. Mater. Sci. 2009, 44, 3006–3009. [CrossRef] 38. Jompang, L.; Thumsorn, S.; On, J.W.; Surin, P.; Apawet, C.; Chaichalermwong, T.; Kaabbuathong, N.; O-Charoen, N.; Srisawat, N. Poly(Lactic Acid) and Poly(Butylene Succinate) Blend Fibers Prepared by Melt Spinning Technique. Energy Procedia 2013, 34, 493–499. [CrossRef] Polymers 2021, 13, 1024 16 of 16

39. Correlo, V.M.; Boesel, L.F.; Bhattacharya, M.; Mano, J.F.; Neves, N.M.; Reis, R.L. Properties of melt processed chitosan and aliphatic polyester blends. Mater. Sci. Eng. A 2005, 403, 57–68. [CrossRef] 40. Chen, R.; Zou, W.; Zhang, H.; Zhang, G.; Qu, J. Crystallization behavior and thermal stability of poly(butylene succinate)/poly(propylene carbonate) blends prepared by novel vane extruder. AIP Conf. Proc. 2016, 1713, 050002. 41. Nattakarn Hongsriphan, K.C.; Kantipongpipat, P.; Thanokiang, J.; Patanathabutr, P. Rheological and Viscoelastic Properties of Poly(Butylene Succinate)/Poly(Ethylene Terephthalate) Blends. In Proceedings of the International Polymer Conference of Thailand: PCT-8, Amari Watergate Bangkok Hotel, Bangkok, Thailand, 14–15 June 2018; pp. 82–88. 42. Roy, M.; Zhelezniakov, M.; de Kort, G.W.; Hawke, L.G.D.; Leoné, N.; Rastogi, S.; Wilsens, C.H.R.M. On the nucleation of polylactide by melt-soluble oxalamide based organic compounds. Polymer 2020, 202, 1–10. [CrossRef] 43. Xia, X.; Weng, G.J.; Zhang, J.; Li, Y. The effect of temperature and graphene concentration on the electrical conductivity and dielectric permittivity of graphene–polymer nanocomposites. Acta Mech. 2020, 231, 1305–1320. [CrossRef] 44. Oi, T.; Shinyama, K.; Fujita, S. Electrical properties of heat-treated polylactic acid. Electr. Eng. Jpn. 2012, 180, 1–8. [CrossRef] 45. Yarin, A.L.; Koombhongse, S.; Reneker, D.H. Taylor cone and jetting from liquid droplets in electrospinning of nanoﬁbers. J. Appl. Phys. 2001, 90, 4836–4846. [CrossRef] 46. Bautista, M.; de Ilarduya, A.; Alla, A.; Muñoz-Guerra, S. Poly(butylene succinate) Ionomers with Enhanced Hydrodegradability. Polymers 2015, 7, 1232–1247. [CrossRef] 47. Zhan, J.; Chen, Y.; Tang, G.; Pan, H.; Zhang, Q.; Song, L.; Hu, Y. Crystallization and melting properties of poly(butylene succinate) composites with titanium dioxide nanotubes or hydroxyapatite nanorods. J. Appl. Polym. Sci. 2014, 131, 1–10. [CrossRef]